EP2735861A1 - Method for determining a transmission value - Google Patents

Abstract

The invention is in the field of in vitro diagnostics and relates to a method for determining a transmission value of a frequency-pulsed light signal through a sample in an automatic analyzer. It further relates to a transmission measuring device for an automatic analyzer, comprising a pulsed with a frequency light source and a photodetector with a downstream A / D converter.

Description

The invention is in the field of in vitro diagnostics and relates to a method for determining a transmission value of a frequency-pulsed light signal through a sample in an automatic analyzer. It further relates to a transmission measuring device for an automatic analyzer, comprising a pulsed with a frequency light source and a photodetector with a downstream A / D converter.

Numerous detection and analysis methods for the determination of physiological parameters in body fluid samples such as blood, plasma, serum or urine or in other biological samples are now carried out automatically in corresponding analysis devices.

Today's analyzers are capable of performing a variety of different detection reactions and analyzes with a variety of samples. Common analytical instruments, such as those used in the clinical laboratory or in blood banks, usually include an area for the supply of sample vessels containing the primary samples to be analyzed. For feeding the sample containers into the analyzer, a transport system is usually provided, which initially transports the sample containers to a sample identification device, which records sample-specific information which is mounted on a sample container and forwards it to a storage unit. Subsequently, the sample containers are transported to a sampling station. With the help of a Probenpipettiereinrichtung there at least an aliquot of the sample liquid is removed from a sample vessel and transferred to a reaction vessel.

The reaction vessels are generally disposable cuvettes, which are kept in stock in a cuvette container in the analyzer and which are automatically transferred from the reservoir into defined receiving positions. The reagents required to provide a variety of assay-specific reactions are contained in reagent containers that are stored in a reagent station. The reagent containers are fed to the analyzer either analogously to the sample vessel feed automatically or manually.

The reagent station usually has a cooling unit to ensure the longest possible shelf life of the reagents. With the aid of a reagent pipetting device, which incidentally often has a heating device, an aliquot of one or more reagents is transferred to a reaction vessel in which the sample to be examined is already located. Depending on the type of biochemical reaction that is initiated by the addition of the reagents to a sample, a different incubation time of the reaction mixture may be required. In any case, the reaction vessel with the reaction mixture is finally fed to a measuring system which measures a physical property of the reaction mixture.

The measurement result is forwarded by the measuring system in turn into a memory unit and evaluated. Subsequently, the analyzer provides a user via an output medium such. As a monitor, a printer or a network connection sample-specific measurements.

Particularly widespread are measuring systems based on photometric (eg turbidimetric, nephelometric, fluorometric or luminometric) measuring principles. These procedures enable the qualitative and Quantitative detection of analytes in liquid samples, without having to provide additional separation steps. The determination of clinically relevant parameters, such as the concentration or activity of an analyte, is often accomplished by simultaneously or sequentially mixing an aliquot of a patient's body fluid with one or more test reagents in the reaction vessel, thereby initiating a biochemical reaction involving a patient measurable change causes an optical property of the test batch. Photometry examines and utilizes the attenuation of a luminous flux during transmission through an absorbing and / or scattering medium. Depending on the nature of the triggered biochemical or biophysical reaction, different photometric measurement methods are used, which enable the measurement of a turbid liquid test batch.

In turbidimetric methods, the turbidity or the optical density of a solution or dispersion (suspension) is measured on the basis of the light attenuation or extinction of a light beam passing directly through the dispersion (suspension). The transmission of the light beam is detected by a photodetector which generates a voltage curve associated with the intensity of the light signal transmitted through the sample.

To improve the signal-to-noise ratio and to suppress low-frequency noise, photodetectors often use a lock-in amplifier. In this case, a modulated measurement signal, ie a transmitted by the sample and received by the photodetector pulsed light signal, fed to a multiplier via a bandpass, which modulates the measurement signal with a phase-locked reference signal of the same frequency and after low-pass filtering generates an output signal which is the amplitude of the modulated input signal is proportional.

Such a lock-in amplifier can also be used in digital systems such. FPGAs (Field Programmable Gate Array), d. H. integrated circuits can be realized. The measurement signal is first fed to an A / D converter, which determines a plurality of samples of the voltage curve from the photometer. The same block structures of an analog circuit are realized by digital blocks. However, it must be to various signal processing units such as bandpass filter, multiplier and low pass generated. Therefore, the system becomes comparatively complex, and an increased amount of logic elements is needed.

It is therefore an object of the invention to provide a method for determining a transmission value and a transmission measuring device of the type mentioned, which have the principle and advantages of the lock-in amplifier in a digital implementation, without causing an increased demand for logic elements.

1. a) Erzeugung einer der Stärke des durch die Probe transmittierten Lichtsignals zugeordneten Spannungskurve,
2. b) Ermittlung einer Mehrzahl von Abtastwerten der Spannungskurve,
3. c) Ermittlung eines einem Lichtpuls zugeordneten ersten Spannungswertes aus einer Anzahl von Abtastwerten und Ermittlung eines einer Dunkelzeit zwischen zwei Lichtpulsen zugeordneten zweiten Spannungswertes aus einer Anzahl von Abtastwerten, wobei die Abtastwerte zur Ermittlung eines einer Dunkelzeit zwischen zwei Lichtpulsen zugeordneten zweiten Spannungswertes gegenüber den Abtastwerten zur Ermittlung eines einem Lichtpuls zugeordneten ersten Spannungswertes um ein ganzzahlig Vielfaches einer der Frequenz des gepulsten Lichtsignals entsprechenden halben Periode versetzt sind, und
4. d) Subtraktion des zweiten Spannungswertes vom ersten Spannungswert.

With regard to the method, the object is achieved according to the invention, in that the method comprises the following steps:

1. a) generating a voltage curve associated with the intensity of the light signal transmitted through the sample,
2. b) determining a plurality of samples of the voltage curve,
3. c) determining a first voltage value assigned to a light pulse from a number of samples and determining a second voltage value associated with a dark time between two light pulses from a number of samples, the samples for determining a second voltage value associated with a dark time between two light pulses with respect to the samples for determining a first voltage value associated with a light pulse by an integer multiple of one of the frequencies of the pulsed one Light signal corresponding to half period are offset, and
4. d) subtracting the second voltage value from the first voltage value.

The difference obtained in step d) is proportional to the transmission value.

The invention is based on the consideration that in the lock-in method in the case of a simply switched signal and reference source in the multiplier ultimately a simple multiplication of the voltage curve determined by the useful signal with the factors +1 and -1 as a function of the phase position the rectangular-modulated signal is carried out. Signal components with useful signal content, d. H. those from light pulses are +1, signal components without useful signal, d. H. of dark times between the light pulses multiplied by -1. If both components contain low-frequency noise components or DC components, they are removed after multiplication by low-pass filtering. The low-pass filtering acts as integrator over several signal periods, so that ultimately takes place a summation on the negative signal-free noise components and the positive signal components. This result is also achieved if a voltage value corresponding to a dark time is subtracted directly from a voltage value corresponding to a light pulse. Thus, a multiplier unit for the signal processing is no longer required.

Sampling values used for the first voltage value and samples used for the second voltage value are offset by an integer multiple of a half period corresponding to the frequency of the pulsed light signal and thus firmly correlated in the phase position. This allows a continuous sampling of the signal or the voltage curve associated with the signal at equal intervals, wherein be selected by the correlation with the frequency and phase of the light pulses, the samples targeted for the signal plus noise component and for the pure noise component. As a result, a continuously low-noise result signal is generated.

In an advantageous embodiment, in step c), i. in the determination of the voltage values, which are assigned to a light pulse or a dark time, samples in a predetermined transition range between light pulse and dark time discarded. This applies to samples that fall in the rising or falling edge of the signal and thus do not give stable signal values. A rejection of such signals improves the accuracy of the result.

Advantageously, the respective dark time, whose sample value is used for the second voltage value, directly follows the respective light pulse, whose sample value is used for the first voltage value, or advances directly to the respective light pulse. On the one hand, a subtraction of successive light pulses and dark periods requires the least technical effort; on the other hand, successive light pulses and dark times are most likely exposed to the same sources of interference, so that a subtraction optimally filters out the sources of interference, which increases the quality of the result.

In a further advantageous embodiment, steps b), c) and d) are repeated cyclically, and an average value is formed via the values determined in step d). The averaging filters additional noise components from the result, so that an even better quality of the result is achieved.

Preferably, a method according to the invention for determining a transmission value in determining the Concentration or activity of at least one analyte in a sample. For this purpose, the concentration or activity of the analyte is determined on the basis of at least one transmission value, which in turn is determined by the method according to the invention.

The invention therefore furthermore relates to a method for determining the concentration or activity of at least one analyte in a sample, wherein the concentration or activity of the at least one analyte is determined on the basis of at least one transmission value which is determined by a method according to the invention. In particular, the invention relates to a method for determining the concentration or activity of at least one analyte in a body fluid sample, preferably in a body fluid sample from the group of whole blood, blood plasma, serum, cerebrospinal fluid or urine.

A transmission measuring device according to the invention comprises means for carrying out the described method.

With respect to the transmission measuring device, the object is achieved in that the A / D converter has a sampling frequency which corresponds to an even multiple of the frequency of the pulsed light signal, and the transmission measuring device has an integrated circuit connected downstream of the A / D converter, which is designed Assign sampling values of the A / D converter on the basis of frequency and phase to a light pulse or a dark time between two light pulses, to delay the samples assigned to light pulses or samples assigned to the dark periods by a half period corresponding to the frequency and thereby to subtract the resulting simultaneous samples from one another.

• i) Abtastwerte des A/D-Wandlers auf Basis der Frequenz der gepulsten Lichtquelle und Phasenlage einem Lichtpuls oder einer Dunkelzeit zwischen zwei Lichtpulsen zuzuordnen, und
• ii) die den Lichtpulsen zugeordneten Abtastwerte oder die den Dunkelzeiten zugeordneten Abtastwerte um ein ganzzahlig Vielfaches einer der Frequenz der gepulsten Lichtquelle entsprechenden halben Periode zu verzögern, und
• iii) dadurch entstehende zeitgleiche Abtastwerte voneinander zu subtrahieren.

The invention therefore further relates to a transmission measuring device for an automatic An analyzer comprising a frequency pulsed light source, a photodetector having a downstream A / D converter with a sampling frequency equal to an even multiple of the frequency of the pulsed light source, and an integrated circuit connected downstream of the A / D converter and adapted therefor is

• i) assign samples of the A / D converter on the basis of the frequency of the pulsed light source and phase position to a light pulse or a dark time between two light pulses, and
• ii) to delay the samples associated with the pulses of light or the samples associated with the periods of dark by an integer multiple of a half period corresponding to the frequency of the pulsed light source, and
• iii) subtract the resulting simultaneous samples from each other.

Advantageously, the sampling frequency has at least four times the value of the frequency of the pulsed light source. As a result, an oversampling is achieved, wherein each light pulse and each dark time at a fourfold sampling frequency already two samples can be assigned.

In a further advantageous embodiment, the integrated circuit is designed to form an average over the subtracted, simultaneous samples. By increasing the digital resolution, especially in conjunction with averaging, the quality of the result signal is further improved.

In a further advantageous embodiment, the integrated circuit includes a Field Programmable Gate Array (FPGA). This is particularly advantageous because in some known automatic analysis devices such FPGAs for processing other tasks, such as for the realization of a processor system, for the evaluation of measurement signals or for controlling motors or various actuators are used and therefore already included in the device. Thus, if an FPGA already exists, additional functions, such as the method described herein, can easily be implemented without the need for redesigning a microprocessor circuit. This increases flexibility over changes in desired new features of an automatic analyzer. Furthermore, it is advantageous that in FPGAs complex functions can be realized at a significantly higher speed than in microprocessors.

Advantageously, an automatic analyzer comprises a described transmission measuring device.

The advantages achieved by the invention are in particular that by using a time shift, subtraction and averaging in the reduction of noise in the photodetector signal of a transmission measuring device simplifying the structure without complex signal processing blocks such. As filter or multiplier is possible. This can cost-effective logic, z. B. simple algebraic operations are used. Not only compared to DSP (Digital Signal Processing) design, but also compared to analog technology, the solution is more cost-effective. There are no complex adjustments such. As gain, phase or filter characteristic necessary. A drift of the signal as with analog methods is also not expected.

FIG 1

eine Transmissionsmessvorrichtung mit einem Lock-In-Verstärker nach dem Stand der Technik,

FIG 2

die prinzipielle Funktionsweise des bekannten Lock-In-Verstärkers aus FIG 1 im Falle eines einfach geschalteten Signals,

FIG 3

eine mögliche Umsetzung der Funktionsweise nach FIG 2, und

FIG 4

einen Ausschnitt einer erfindungsgemäßen Transmissionsmessvorrichtung mit einer korrelierten Überabtastung.

The invention will be explained in more detail with reference to a drawing. Show:

FIG. 1

a transmission measuring device with a lock-in amplifier according to the prior art,

FIG. 2

the basic operation of the known lock-in amplifier FIG. 1 in the case of a simply switched signal,

FIG. 3

a possible implementation of the functionality FIG. 2 , and

FIG. 4

a section of a transmission measuring device according to the invention with a correlated oversampling.

Identical parts are provided with the same reference numerals in all figures.

The transmission measuring apparatus 1 according to the FIG. 1 shows the operation of a conventional lock-in amplifier. It initially has a frequency generator 2, to which a modulator 4 is connected downstream. The modulator 4 modulates the signal, which is typically oscillated by a zero value, of the frequency generator 2 such that a uniform turn-on and turn-off oscillation for a light source 6, in this case a light-emitting diode, is achieved. The light source 6 is thus pulsed with the frequency of the frequency generator 2. Alternatively, a continuous light source 6 with a chopper wheel with corresponding rotational frequency could also be used.

The light source 6 irradiates a sample 8, in the present case a blood sample in a not shown automatic analyzer. The blood sample is subjected to a chemical reaction, during which the absorption of the light of the light source 6 is to be determined by means of a transmission measurement. The transmitted light is detected by a photodetector 10.

The photodetector 10 is first followed by a bandpass filter 12 whose transfer function has a maximum in the range of the frequency of the frequency generator 2. On the one hand, the bandpass filter 12 filters out interfering signal components outside the pulse frequency of the light source 6, on the other hand it again generates a signal oscillating around zero from the signal oscillating between zero and a maximum value of the photodetector 10. This is multiplied in a multiplier 14 with the signal from the frequency generator 2, which is phase-shifted in a phase shifter 16. The phase shift is adjusted in such a way that a maximum output signal is achieved at the multiplier 14.

The output signal is finally fed to a low-pass filter 18, which ultimately acts as an integrator and outputs a DC voltage signal as a result signal 20. The result signal 20 is proportional to the transmission through the sample 8 and largely free from interference. This in FIG. 1 In particular, the method shown is used to increase the sensitivity for low signal levels. Influences of low-frequency noise (eg 1 / f noise) and extraneous light (daylight, artificial lighting) are reduced.

An implementation of in FIG. 1 shown transmission measuring device 1 in digital systems such. As FPGAs is basically possible, but requires a relatively complex digital implementation of DSP modules such as bandpass filter 12, multiplier 14 and low-pass filter 18th

FIG. 2 shows a simplified representation of the operation of the multiplier 14 in the case of a simple switched signal such. B. a pulsed light source 6 in an automatic analyzer. A phase shift is typically not expected. The photodetector signal 22 is multiplied by the square-wave signal 24 oscillating between -1 and +1 from the frequency generator 2. Since the + 1 regions always correspond to the light pulses of the light source 6, thus signal components with useful signal S are always multiplied by +1, while signals from dark periods, which contain only noise N, are multiplied by -1. This results in the FIG. 2 represented output signal 26 with alternating S + N and -N regions. If both components contain low-frequency noise components or DC components, they are removed after multiplication by low-pass filtering.

In fact, this can be done in FIG. 2 principle shown by the in FIG. 3 implement shown circuit. The photodetector signal is split into two branches 28, one of which is multiplied by -1. The corresponding signals are to each branch 28 in FIG. 3 shown. The square-wave signal 24 controls a switch 30, which switches with the frequency of the rectangular signal 24 between the two branches 28. This creates the already in FIG. 2 shown output signal 26th

Subsequent low pass filtering is simple, in FIG. 2 and 3 case illustrated as averaging over several periods feasible. In the process, a subtraction of the noiseless noise N from the useful signal with noise component S + N is thus ultimately carried out. It results as an output signal S = (S + N) -N.

However, this result is also achieved if the signal-free noise component N is subtracted directly from the preceding or following signal component S. This is no Multiplier 14 more required. A transmission measuring device 1 operating according to this principle is shown in FIG FIG. 4 shown in detail. In the FIG. 4 The circuit shown is implemented on an integrated circuit, namely an FPGA.

The photodetector signal 22 is supplied to an A / D converter 31, which performs scanning of the S + N regions as well as the N regions on the one hand. In the exemplary embodiment, the A / D converter 31 is designed as an 18-bit A / D converter. The frequency of the light source is 45 kHz, the A / D converter is designed for a 20-times sampling frequency, so that each ten samples 32 per period are stored and processed as S + N samples 34 and the N samples 36. Since the frequency of the A / D converter 31 is correlated with the frequency of the light source 6, the samples 32 are immediately attributable to a light pulse or a dark time. Samples 32 that fall on the rising or falling edge are ignored or discarded.

By oversampling a lock-in period of twenty signal values in the exemplary embodiment using in each case four signal component values and four noise component values, the result of the signal value calculation is already increased by a factor of four, which corresponds to two bits of digital resolution. Uncorrelated high-frequency noise is reduced by a factor of four root, which leads to an initial improvement of the signal-to-noise ratio.

The S + N samples 34 are delayed in a delay element 38 by a number of samples 32 corresponding to a half period of the frequency of the light source 6. In a subtracter 40, simultaneous S + N delayed samples 34 and N samples 36 are subtracted from each other. As a result, in particular a bandpass filtering connected downstream of the photodetector 10 can also be dispensed with. The output signal 42 of the subtractor 40 is a simple, over N periods adjustable averaging 44 supplied. In the exemplary embodiment, over 1024 lock-in periods, ie periods of the light source 6, are averaged.

The circuit shown results in a corresponding increase in the resolution of 10 bits for the result signal 20 and, for uncorrelated noise, a further improvement in the signal-to-noise ratio of theoretically root 1024 = 32 (or 30 dB, 5-bit digital). Under real conditions, however, no uncorrelated noise can be assumed, so that the real resolution gain is lower. Measurements have resulted in an equivalent resolution of about 22 to 23 bits in the embodiment, which corresponds to a signal to noise ratio of more than 130 dB. The sensitivity to externally impinging ambient light is very low.

LIST OF REFERENCE NUMBERS

1

Transmission measuring device

2

frequency generator

4

modulator

6

light source

8th

sample

10

photodetector

12

Bandpass filter

14

multipliers

16

phase shifter

18

Low Pass Filter

20

result signal

22

Photodetector signal

24

square wave

26

output

28

branches

30

switch

31

A / D converter

32

samples

34

S + N samples

36

N samples

38

delay

40

subtractor

42

output

44

Averaging calculation

S

payload

N

sough

Claims (11)

Method for determining at least one transmission value of a frequency-pulsed light signal through a sample (8) in an automatic analyzer, comprising the steps of: a) generating a voltage curve (22) associated with the intensity of the light signal transmitted through the sample (8), b) determining a plurality of samples (32, 34, 36) of the voltage curve (22), c) determining a first voltage value associated with a light pulse from a number of samples (34) and determining a second voltage value associated with a dark time between two light pulses from a number of samples (36), the samples (36) determining a dark time between two Light pulses associated with the second voltage value relative to the samples (34) for determining a light pulse associated with a first voltage value by an integer multiple of the frequency of the pulsed light signal corresponding half period are offset, and d) subtracting the second voltage value from the first voltage value,
wherein the difference obtained in step d) is proportional to the transmission value. The method of claim 1, wherein in step c) samples (32) are discarded in a predetermined transition range between light pulse and dark time. Method according to Claim 1 or 2, in which the respective dark time directly follows the respective light pulse or immediately precedes the respective light pulse. Method according to one of the preceding claims, in which steps b), c) and d) are repeated cyclically and an average value is formed via the transmission values determined in step d). A method for determining the concentration or activity of at least one analyte in a sample characterized in that the concentration or activity of the at least one analyte is determined using at least one transmission value, which is determined by a method according to any preceding claim. The method of claim 5, wherein the concentration or activity of the analyte in a body fluid sample, preferably in whole blood, blood plasma, serum, cerebrospinal fluid or urine is determined. Transmission measuring apparatus (1) for an automatic analyzer, comprising a frequency-pulsed light source (6), a photodetector (10) having a downstream A / D converter (31) with a sampling frequency equal to an even multiple of the frequency of the pulsed light source , and an integrated circuit connected downstream of the A / D converter (31), which is designed to i) assign sampling values of the A / D converter (31) based on the frequency of the pulsed light source and phase position to a light pulse or a dark time between two light pulses, and ii) delaying the samples (34) associated with the pulses of light or the samples (36) associated with the periods of darkness by an integral multiple of a half period corresponding to the frequency of the pulsed light source, and iii) subtract from each other thereby resulting simultaneous samples (32). A transmission measuring device (1) according to claim 7, wherein the sampling frequency has at least four times the value of the frequency of the pulsed light source (6). A transmission measurement device (1) according to claim 7 or 8, wherein the integrated circuit is adapted to form an average over the subtracted, simultaneous samples (32). Transmission measuring device (1) according to one of Claims 7 to 9, in which the integrated circuit contains a Field Programmable Gate Array (FPGA). Automatic analyzer with a transmission measuring device (1) according to one of claims 7 to 10.

Images